AI AI in Teaching and Learning

An AI practitioner’s guide

Clear expectations and responsible AI use in college courses

Understand AI-use options, communicate expectations to students, and connect course choices to institutional guidance.

  1. Step 1 Understand AI use → AI and learning
  2. Step 2 Choose and communicate expectations → Policy builder
  3. Step 3 Identify responsible-use considerations → Principles
  4. Step 4 Connect to institutional guidance → Policy artifacts
  5. Step 5 Find sources and support → Evidence

What this guide is

This independent resource offers an AI practitioner’s perspective on AI use in higher education. It helps faculty, students and institutional leaders understand AI-use options, clarify expectations and consider responsible-use and governance implications. It does not prescribe teaching methods, validate assessment designs, or replace disciplinary expertise, instructional-design support or institutional policy.

What do you need to do?

Short on time? The policy builder turns your choices into an assignment-level AI statement you can adapt.

Build an assignment AI policy →
Write an assignment AI policy → Policy builder Redesign an assessment → AI considerations for assignments Understand student expectations → For students Develop institutional guidance → Guideline types and frameworks

Background in four figures

180 / 200

top-ranked universities publish GenAI guidelines (2026)

Website review, April 2026. Shows guidelines exist, not that they work. Source and method ↗

94%

of UK undergraduates use AI for assessed work (HEPI, 2026)

UK survey. Measures any use, not how often or how well. Source and method ↗

40–45%

of syllabi in a 31,000-syllabus study still lack an AI policy (2026)

Syllabi from 2021–2025. Working paper, not yet peer reviewed. Source and method ↗

5

assessment-design options on the AI Assessment Scale, not allow-or-ban

A framework, not a research finding. Source and method ↗

Sources: Top 200 universities study, HEPI, Chirikov, AI Assessment Scale.

The AI Assessment Scale at a glance

Open the policy builder →
Level 1 No AI Students work without AI, often in supervised settings. Works well for: foundational skills, in-class writing, exams and tasks where independent ability is what you assess.
Level 2 AI Planning AI for brainstorming, outlining and finding sources. Drafting is the student’s own. Works well for: research projects and essays where idea generation is not the main learning goal.
Level 3 AI Collaboration AI helps draft and revise. Students evaluate, edit and disclose. Works well for: writing-intensive and project courses that teach students to work critically with AI.
Level 4 Full AI AI used throughout. Students direct it and stay accountable. Works well for: professional practice tasks, advanced projects and assignments where using AI well is part of the skill.
Level 5 AI Exploration The task is about AI: experimenting with it, testing it or critiquing it. Works well for: AI-literacy units, design and computing courses, and assignments that evaluate AI itself.

These are assessment-design options, not levels of teaching quality. Pick the option that fits the learning outcome and the evidence students must produce. One course can use different levels for different assignments, or for different stages of one assignment.

Levels from the AI Assessment Scale (Perkins, Furze, Roe & MacVaugh).

Start where you are

Three routes through the guide.

Faculty

Set AI expectations for a course or an assignment.

AI and learning → Policy builder → Syllabus elements → Glossary →

Students

Understand what instructors expect, how to disclose AI use, and your rights if work is flagged.

For students: disclosure and your rights → What syllabus policies cover → Why detector scores are not proof →

Academic leaders

Set institutional guidance, governance and faculty support.

Policy artifacts → Frameworks → Evidence →

Downloads

Editable files, in the order you would use them.

  1. Step 1Decide what students must demonstrate.
  2. Step 2Turn that decision into explicit rules.
  3. Step 3Make permitted assistance visible.
  4. Step 4Introduce, apply and revisit the policy.

The guide, page by page

Read in order or jump to a page.

Principles Six principles, each grounded in recent sources AI and learning AI roles, offloading research and AI literacy AI considerations for assignments Six questions and who decides Guideline types Eight kinds of guidance, with real examples and evidence AI detectors The past year’s evidence, and how campuses respond For students Disclosure, flagged work and your rights Policy builder Five levels of AI use, draft statements and real examples Syllabus elements What published syllabus statements have in common Policy artifacts Seven layers and ten artifact types, applied to teaching Frameworks Current frameworks and the three tiers of policy Evidence Twelve recent studies, plus foundational and superseded sources Glossary Key terms, defined References Every source cited, with links About Author, scope, methods and contact